Coalition formation for multiple heterogeneous UAVs cooperative search and prosecute with communication constraints

Abstract

To improve the cooperative search and attack effectiveness of multiple heterogeneous unmanned aerial vehicles (UAVs) in unknown environment, a novel coalition formation method with communication constraints is presented in this paper. First, the coalition formation model is established on the basis of minimizing the target attack delay and minimizing the coalition size with the constraint of required resources and simultaneous strike. Second, considering communication constrains such as limited communication ranges and communication delays, a mechanism was developed in order to find the potential coalition members within a maximum number of hops over a dynamic UAV network. Third, to reduce the huge computational complexity in coalition formation optimization solution we propose a Multistage Sub-Optimal Coalition Formation Algorithm (MSOCFA) with low computational complexity. Furthermore, in order to enable multiple cooperative UAVs accomplish the search and prosecute missions autonomously, a distributed autonomous control strategy is proposed which is based on the Finite-State Machine (FSM). Comparison simulations are carried out to demonstrate that how the potential coalition members finding technique impact on the coalition achieved by MSOCFA. The effects of number of maximum allowed hops for a message and hop delay are studied by employing Monte-Carlo method. The experimental reveals that, in the cases of large communication delay, forming a coalition from the immediate neighbors is sufficient for a good performance in term of the mission completion time. Under smaller delays, including neighbors up to a few hops will increase performance, and any additional increase in hop count will degrade performance.